Ž .
Journal of Health Economics 19 2000 679–695 www.elsevier.nlrlocatereconbase
Health care expenditure in the last months of life
Stefan Felder
a,
, Markus Meier
b
, Horst Schmitt
a
a
Health Economics Department, Faculties of Medicine and Economics, Otto-Õon-Guericke UniÕersity Magdeburg, Leipziger Strasse 44, 39120 Magdeburg, Germany
b
Department of Economics, Socioeconomic Institute, UniÕersity of Zurich, Zurich, Switzerland Received 12 May 1997; received in revised form 11 May 1999; accepted 6 January 2000
Abstract
Ž .
In OECD countries, a considerable share of health care expenditure HCE is spent for the care of the terminally ill. This paper derives the demand for HCE in the last 2 years of
life from a model that accounts for age, mortality risk and wealth. The empirical tests are based on data of deceased members of a major Swiss sick fund. The empirical evidence
Ž . confirms most of the hypotheses derived from the model, i.e., i HCE increases with
Ž . Ž .
closeness to death, ii for retired individuals, HCE decreases with age, and iii low-income individuals, as compared to high-income individuals, incur lower HCE in the last months of
life. q 2000 Elsevier Science B.V. All rights reserved.
JEL classification: I12; I31; D61 Keywords: Mortality risk; Willingness to pay for survival; Demand for health care services
1. Introduction
The cost of treating the terminally ill is sometimes blamed for the steady increase in health care costs. Indeed, the so-called cost of dying is substantial. As
Ž .
Lubitz and Riley 1993 report for the United States, 30 of the total Medicare
Corresponding author. Tel.: q49-391-532-8050; fax: q49-391-671-90250. Ž
. E-mail address: stefan.feldermedizin.uni-magdeburg.de S. Felder .
0167-6296r00r - see front matter q 2000 Elsevier Science B.V. All rights reserved. Ž
. PII: S 0 1 6 7 - 6 2 9 6 0 0 0 0 0 3 9 - 4
budget is paid out on behalf of persons in their last year of life. Spending per decedent is about seven times larger than per survivor. For Switzerland, health
Ž .
care expenditure HCE of the terminally ill is comparably high, as samples of two Ž
. health insurance companies show cf. Zweifel et al., 1999 . Payments for persons
in their last year of life constitute 22 and 18 of the total health care cost of retired persons, depending on the sample. The average ratio of per capita expendi-
ture between decedents and survivors equals 5.6:1.
1
Besides its size, the path of HCE in the last year of life is exceptional as well. One observes a steady increase, accentuated by a surge in the last quarter of life.
Ž .
In the study by Lubitz and Riley 1993 , 62 of all Medicare costs referring to the last year of life were incurred in the last quarter, compared to 10 in the first
Ž quarter. In Switzerland, the difference is less accentuated 42–48 as opposed to
Ž ..
14–18, depending on the sample cf. Zweifel et al., 1999; Table 1 . A third characteristic of HCE in the last months of life is age dependency.
Ž .
Baker et al. 1995 reported that cost in the last 2 years of life for persons who died at 101 years of age or older was 37 of those incurred by persons who died
at 70. In the Swiss samples, we found a similar pattern with respect to age and HCE in the last 2 years of life.
The present paper presents a model designed to explain these three character- istics of HCE in the last months of life. It derives the demand for health care
